© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21692
    Prev
    Next
    INDEX
    Explanations

    not what I want

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    满æĦıçļĦ
    -0.10
    ä¸İåIJ¦
    -0.09
    æĽ´åħ·
    -0.09
    è¾ĥå°ij
    -0.09
    æīįå¹²
    -0.08
    IVATE
    -0.08
    éģ¿å¼Ģ
    -0.08
    满æĦı度
    -0.08
    æĹ©æĹ¥
    -0.08
    enden
    -0.08
    POSITIVE LOGITS
    abi
    0.08
    ophage
    0.08
    ä¹ħ
    0.08
    è¶Ĭ
    0.08
    å¨ģ
    0.08
    İ
    0.08
    .sparse
    0.08
    âĸĮ
    0.07
    羯
    0.07
    ONY
    0.07
    Activations Density 0.088%

    No Known Activations